Blog is co-authored by PRASANNA KELKAR

Introduction

As IBM released latest 2.3 release of IBM Spectrum fusion in September 2022, Overall many new features are added and enhanced the performance including many features related Serviceability and Logcollection of IBM Spectrum fusion. This article provides a overview of features enhanced in 2.3 release of IBM Spectrum fusion focused on Serviceability.

What are new features :

Most of the features added are aligned with the customer needs keeping the focus on simplifying logcollection process in various use cases. These enhancements have streamlined the logcollection process in such a way that when issue occurs, logs would get collected automatically as part of fusion logs itself instead of collecting logs through manual steps separately.

For e.g. Just to name few: IBM Spectrum Scale must gather logs inclusion or IBM SPP agent and server logs inclusion etc.

Lets understand more details below.

How it helps in solving the problem

Prior version does provide option to collect logs but its expanded to have more comprehensive logs in latest version. Currently, logs get collected either through a manual step via IBM Spectrum fusion User interface or through automated way while creating new IBM support ticket for critical event. Either way, It will have 6 predefined packages which pretty much covers all logs required by IBM support team. So new logs files are included in these existing packages itself.

Following UI menu allows user to collect logs under these predefined 6 packages as of now.

Logcollection Features

The new logs added in various log collection packages, are part of a continuous effort with the aim to enrich the existing collection set and to make it more relevant to the actual field scenarios. System administrator or IBM support team might refer these log files while debugging  variety of issues like any component failure in IBM Spectrum fusion or issues related to IBM Spectrum Protect plus or IBM Spectrum scale or any Node etc.

Following logs are newly added.

1. IBM Spectrum Scale must-gather logs are now added in Storage log collection.

   These are OCP must-gather logs which are specific to IBM Spectrum Scale and helpful for debugging its issue related IBM Spectrum Scale.

2. IBM Spectrum Protect Plus agent and server logs are included in backup and restore log collection. This would be prerequisite for administrator or support for debugging any issues related to backup and restore component of Spectrum fusion.

3. OCP must-gather logs are now included in Administration log collection

   This contains information from your cluster that is most likely needed for debugging issues, including but not limited to resource definition and service logs.

4. GPFS snap logs are added in Storage log collection

   Creates an informational system snapshot at a single point in time. This system snapshot consists of information such as cluster configuration, disk configuration, network configuration, network status, GPFS logs, dumps, and traces.

5. isf-operator’s ClusterServiceVersion file is now included by default in every log collection request.

   This file is useful for identifying versions of different dependent components (operators and microservices). This could be helpful to debugging.

Apart from the new log additions mentioned above here are few improvements in Serviceability area .

1. New Collection sets defined for SDS (Software Defined Storage Service)
   Now the log collection will contain more relevant data which will be helpful in debugging SDS issues.

2. Informative filenames for downloaded zip files
   Now the downloaded zip file will have more meaningful name. It will comprise of two parts. A short form of the log collection name followed by a timestamp when this log was requested.So user can guess the time when logs are extracted.
   e.g. br-20221115091230.zip represent backup and restore package logs taken at 15 Nov 2022

3. Improvement in log collection time for IMM logs.

   With this release overall time required to download all IMM logs is reduced by around 50%(approx). As a result of this Network log collection which contains these logs will also see the improvement.

Event manager features.

If any part of the fusion goes down usually corresponding events will be shown in fusion User Interface but if for some reason EventManager microservice itself goes down , an OpenShift alert will be raised to notify the administrator.

Callhome features

• In case of failure of callhome in creating a ticket for any of a observed issue, a fusion event with critical severity will be raised. This will be help for user since they will be alerted about the callhome issues in time.

• Fresh new screen is introduced in Support logs page, that help use to attach logs to existing ticket or create new ticket. User can now manually upload previously collected logs onto an existing callhome ticket or automatically create a new ticket and upload these logs.

  

Conclusion:

Overall all these enhancements should effectively help the administrators in making sure, system in running smoothly and in case of a failure, it is handled appropriately.

Acknowledgement

Sincere thanks to Shyamala Rajagopalan for helping review the article.